Sequence Comparative Analysis using Networks (SCAN) is a software developed for evaluating de novo transcript assemblies build on next generation sequencing data.
SCAN is designed to compare distinct node sets in networks, to make comparisons accross assemblies and for statistical comparisons between distinct assemblies. The strength of SCAN is providing statistical support for transcript assemblies in a biological context.
In an evaluation of SCAN, the software is tested on three transcriptomes assembled using three different assembly programs: Velvet, ABySS and CLC.
De novo transcriptome assemblies of the non-model oomycete Achlya hypogyna produced by ABySS, Velvet and CLC Genomics Workbench (CLC) assembly algorithms are evaluated using the SCAN software. Predicted proteins from the fully sequenced and annotated Pythium ultimum genome are used as the reference dataset to assess the significant similarity of transcripts reconstructed from the A. hypogyna mRNA data to known biological sequences.
The evaluation is testing efficiency in terms of “good” transcripts in a number of different aspects as e.g. the number of good transcripts from the assembly based on supported similarity to reference data, local articuation point test, length of chains (quality transcripts) and Jaccard Index. Consistently, CLC assembly – compared to ABySS and Velvet – results in the highest number of quality transcripts.Read the interesting article
Sequence Comparative Analysis using Networks (SCAN): software for evaluating de novo transcript assembly from next generation sequencing
Ian Misner, Cédric Bicep, Philippe Lopez, Sébastien Halary, Eric Bapteste, and Christopher E. Lane
Mol Biol Evol first published online May 10, 2013 doi:10.1093/molbev/mst087